Core Web Vitals in 2026: INP is the new fight, here is the playbook

INP (Interaction to Next Paint) replaced FID as a Core Web Vital in March 2024. The two-year grace period is over, and Google’s ranking signal is now using it. Yet 6 of 10 sites we audit in 2026 still have INP over the 200ms “good” threshold.

Why INP is harder than FID#

FID measured only the first input. INP measures every interaction over a session and reports the worst (or 98th percentile). One janky click during a session can torpedo the score.

The five fixes#

1. Break up long tasks. Any synchronous JS over 50ms blocks the main thread. Audit with Chrome DevTools Performance panel, split with scheduler.yield() or setTimeout(0).
2. Defer non-critical interactivity. The fourth-tab dropdown does not need to mount on first paint. Use dynamic import + Intersection Observer.
3. Use CSS for hover/focus states. If a JS event handler updates a class, the CSS could have done it directly. Saves a re-render.
4. React: useDeferredValue for expensive re-renders. Search results, filter panels. Lets React keep the input responsive while the heavy list updates.
5. Audit third-party scripts. Most chat widgets, A/B test snippets and tag managers spawn long tasks. Either lazy-load them or replace with lighter alternatives.

Measure on real users#

Synthetic Lighthouse INP is a rough guide. CrUX (Chrome User Experience) data in Search Console is the truth. Look at the p75 INP for your top pages; that is what Google ranks you on.

We bake INP-aware patterns into every build from day one. Audit your INP if your traffic dropped after March 2024.